Which timeframe is best for SmallCatZzFractal v1.1 ??
ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies
Alert on Fractals (MT) 2 replies
Zigzag & Fractals 10 replies
EA wanted for sound alert depending on fractals condition 2 replies
Fractals Edge Fx 3 replies
Disliked{quote} i think you are mistaken that indi you are asking it tf is not the one that gg53 and other are u using as non repainting.so that one from small catzfractal is meant to much the fractal and zz endpoint but it repaint.just start reading from ultimate thread first and then follow this thread after.Ignored
DislikedI'm having very similar results, Burnssss. Have you calculated what the overall percentage roughly is for great signals? If I look at all my signals, I see that those bad ones start quite early and continue right to the turning point and then fade. At the moment I've got no idea how to balance this issue. Given the data and logics, it is not misbehaving. Testing notes on the new "most probable length" - it's not quite the same as the quasi-G, 'cause 4th bar (length+4) is used, but the results are as I expected - too trigger-happy. Upside is it responds...Ignored
Disliked{quote} Currently testing the trigger of 65% "in-length" (-3 to 0) / (0 +3) And 85% "in-zone" (-3 to +3) / ( higher than +3) Carl @candero, I see the same sometimes... several signals in a row, the last signal being a great one for the turn.Ignored
DislikedHi G, When you first ran the indicator with out of sample data at the start of 2013, i.e. the adaptation algo hadn't done much to update the database yet... Were the predictions almost as accurate and complete (relative to the number of ZZ turning points) as they are now? What was the win rate/profitability estimated to be compared to now? When you check the win rate/profitability now, is it always after running the adaptation algo for several months first? What win rate should we try to aim for when we don't have the Monty Hall or adaptation algo?...Ignored
DislikedI finished (?) testing pseudo real-time updating of the model (i.e. constant updating of historical data). As I thought, it causes two "bad" side effects: 1. On (rare) ocassions it causes "repainting" or double signals if the previous one stays on screen. -->> B A D 2. In some more cases it causes the logic of alternating signals (which is my base point and purpose) to "fail" - i.e. two sequences of UP's or two DOWN's -->> B A D G.Ignored
DislikedWhy wouldn't the actual ZZ leg, pattern, and lengths of signals 4, 5, 6, and 7 be entered into the database? Those ones all occurred within one MT4 ZZ.Ignored